Hello! We’ve been working to build a better foundation for collecting and representing usage of various Fly.io resources so we can all better reason about our apps, why they’re costing money.
We want you to be able to more easily see how many VMs of a certain size were running last week - at the app or organization level, see how many gigabytes of volume were provisioned last tuesday, or how many public IPs you had over the month.
We aren’t all the way there yet, but we’ve recently put a lot of the basic pieces in place to provide these better tools.
Today, we’re exposing some tracking we’ve started doing for provisioned IPs.
These are still pretty experimental while we work out some of the kinks of how we’re tracking this data and showing it back to you. They’re so experimental that your graphs probably show you have data for Friday, but then not again until this past Monday.
This is a baby step towards comprehensive billing data. People building with Fly Machines need very granular billing data so they can account for their per customer usage costs. We generate a ton of metrics, so getting these into a good UX is actually pretty hard, because we first need big data .
IPs aren’t a big data problem (yet), so they’re a nice place to start, but we’re hoping to add everything that might cost you money over time.